Comparison of the Thermal Effects on LWIR Optical Designs Utilizing Different Infrared Optical Materials
Total Page:16
File Type:pdf, Size:1020Kb
Invited Paper Comparison of the thermal effects on LWIR optical designs utilizing different infrared optical materials Jeremy Huddleston, Alan Symmons, Ray Pini LightPath Technologies, Inc., 2603 Challenger Tech Ct, Ste 100, Orlando, FL, USA 32826 ABSTRACT The growing demand for lower cost infrared sensors and cameras has focused attention on the need for low cost optics for the long wave and mid-wave infrared region. The thermal properties of chalcogenides provide benefits for optical and optomechanical designers for the athermalization of lens assemblies as compared to Germanium, Zinc Selenide and other more common infrared materials. This investigation reviews typical infrared materials’ thermal performance and the effects of temperature on the optical performance of lens systems manufactured from various optical materials. Keywords: Precision glass molding, long wave infrared, LWIR, thermal camera, athermal, athermalization, chalcogenide, SWaP-C. 1. INTRODUCTION Over the past ten years, the prices for thermal imaging sensors have decreased dramatically. The resulting cost savings has significantly increased the quantities of thermal imagers sold. However, the high cost of traditional materials for thermal lenses has become the limiting factor in overall camera costs. Optics manufacturers are looking for high volume, low cost methods to produce infrared optical systems. The technological roadmap for infrared optical systems is following the same path that visible camera systems have followed in the recent past; from large SLR type cameras to small handheld cameras and finally to cell phone camera systems. The enabling technology for the visible region was molding of optical lenses. The advantages of molded optics apply even more so to the LWIR region, where the potential for cost reduction relative to existing lens technologies is even greater.1 Specifically, the chalcogenide moldable materials for IR have improved thermal properties relative to traditional IR materials, such as Germanium.2 These athermal characteristics are extremely valuable in reducing the cost, size and complexity of mechanical compensation techniques required for less athermal materials. Furthermore, the availability of different commercial chemical compositions of the chalcogenides enable potential trade offs on optical and mechanical characteristics. The focus of this study is to explore the potential performance benefits of these moldable materials, and additionally to improve upon the thermal analysis techniques used to evaluate the expected thermal effects on system performance for low-cost optics. 2. ATHERMAL DESIGN THEORY 2.1 Theory Introduction There are three main ways to combat the effects of temperature in an optical system: active mechanical compensation (electromechanical), passive mechanical compensation (optomechanical), and passive optical compensation (material).3 Each successive method represents reduced cost, size and complexity. Most lenses may be mechanically compensated to some degree, so in order to reduce system size and cost, the lens material properties must provide most or all of the compensation. Therefore, we will address the effects of passive optical athermalization in the absence of mechanical compensation, by comparing the properties of various lens materials. Ideally, an “athermal” or “athermalized” lens would have no change to image quality over the full operating temperature range across the full FOV of any detector with which it is used. In practice, there are many limitations to this ideal for which a lens user or purchaser should be aware. Infrared Technology and Applications XL, edited by Bjørn F. Andresen, Gabor F. Fulop, Charles M. Hanson, Paul R. Norton, Proc. of SPIE Vol. 9070, 90702E · © 2014 SPIE CCC code: 0277-786X/14/$18 · doi: 10.1117/12.2050686 Proc. of SPIE Vol. 9070 90702E-1 The first issue is that there is no agreed-upon quantitative definition for an athermal lens, such that lens providers may choose to call any lens “athermal” without specifying the actual lens performance over temperature4. The athermal characteristics of a lens system may be evaluated in various ways, ranging from theoretical approximations (“rules of thumb” analytical formulas) to predictive modeling of expected performance through rigorous optomechanical design of the lens, housing and detector components. In some cases, the use of the term “athermal” simply describes a decrease in thermal sensitivity relative to other lenses. Such a lens may also be described as having “material athermalization”, meaning that the natural thermal properties of the lens material make it relatively insensitive to temperature compared to traditional materials and existing lenses on the market. This leaves much ambiguity as to the actual level of athermal performance that one should expect when using the lens. Another limitation is that ideal athermalization techniques are often complex and expensive. With growing industry pressure to reduce the cost and size of lens systems, many of these theoretical methods for athermalization have become less practical or affordable. Optically passive techniques require multiple elements with varying materials, which increase cost and size. Mechanically (or optomechanically) passive techniques require multiple materials in the lens or camera housing which may also increase the cost and size of overall camera system. Thus, a trade-off must be found between an ideal solution for perfecting image quality over temperature, and achieving a small, low-cost camera system. Further limitations arise from the incomplete use of optical theory to implement athermalization. Early efforts in deriving formulas for athermal solutions relied on many approximations and assumptions in the underlying optical theory (ex. paraxial optics, thin lenses, no aberrations, ideal housings, etc).3,5,6,7 These approximations are often misunderstood, and the assumptions are often unrealistic when applied to practical design problems. The impact can be even more problematic when trade-offs of performance and material selection are necessary for achieving aggressive goals of cost and size. For instance, small focal lengths, low F numbers, large FOVs, ray aberrations, housing constraints, and detector design can all lead to reduced validity of the theoretical formulations. The aforementioned lack of an athermal image quality standard has enabled these approximations to persist in literature and replace a more a rigorous methodology for athermalization analysis. 1,3,4,8,9,10 The scope of this study is to begin with the well-known analytical approximations, and then compare to more realistic expectations from modeling of practical systems. This will form the proper foundation for comparing a select group of available IR lens materials to reveal advantages that can be utilized for achieving realistic design goals. 2.2 Predictions of Analytical Theory Athermalization theory has been well documented and has been used for athermalized lens designs, often without addressing the applicability of the underlying assumptions. Since we wish to establish a baseline relative to broadly used theory, we will review this theory first without addressing the weaknesses of these assumptions, and later compare to modeled results. We start by noting that the change in focus Δf with temperature ΔT of an ideal lens of nominal focal length fo may be calculated by the use of a thermal glass constant γT related to the refractive index n, the temperature coefficient of 4,8,10 refractive index dn/dT, and the coefficient of thermal expansion (CTE) of the lens material αL as follows: ∆ ∙ ∙∆ (1) where ∙ (2) This only addresses the lens, so we can add a housing CTE αH to equation (1) to obtain a combined thermal constant ßT, (3) where the thermal focal shift now becomes ∆ ∙ ∙∆ (4) This treatment assumes that the housing mounts to the lens at the principal plane and mounts to the detector at the image plane, as shown in Figure 2 below. Proc. of SPIE Vol. 9070 90702E-2 Figure 1 – Simplified optomechanical assumptions of equation (4) from approximate athermalization theory. This is an erroneous assumption in most cases, but if we temporarily accept it, we find that the change in focus is proportional to the sum of the housing CTE and the thermal glass constant. The thermal properties referenced above for a few conventional IR materials and several formulations of chalcogenides are found in Table 1, with the assumption of an Al housing. Although the chalcogenide materials selected may be obtained under various trade and brand names, we will refer to them by their chemical composition to avoid confusion. Table 1: Optical/thermal properties of several LWIR lens materials 1 2 3 4 4 4 Property Symbol Unit Ge ZnSe ZnS Ge10As40Se50 Ge28Sb12Se60 As40Se60 Index at 10.6µm n N/A 4.003 2.403 2.192 2.609 2.600 2.778 Thermal Coef. of dn/dT 10-6/°C 400 61 43 20 70 32 Refractive Index -6 Lens CTE αL 10 /°C 5.9 7.1 6.6 20.4 14.5 20.8 Thermal Glass -6 γT 10 /°C 127 36 29 -8 29 -3 Constant -6 Housing CTE (Al) αH 10 /°C 23.6 23.6 23.6 23.6 23.6 23.6 Combined -6 ßT 10 /°C 151 60 53 16 53 21 Thermal Constant 1http://eom.umicore.com/en/infrared-optics/blanks/germanium-datasheet.pdf 2http://www.crystran.co.uk/userfiles/files/zinc-selenide-znse-data-sheet.pdf 3http://www.crystran.co.uk/userfiles/files/zinc-sulphide-flir-zns-data-sheet.pdf 4LightPath Technologies, measured at 10.6µm wavelength The thermal focal shift we have described is only meaningful when applied to a relevant criterion for acceptable focal shift that keeps the lens “athermal”. As previously mentioned, there are no well-established performance criteria, so many have chosen to use the “diffraction limited depth of focus” DOFDL of the lens, which is a theoretical value derived by further approximations to be: 4,8,10 2∙λ∙/# (5) By this definition, the lens is “athermal” if the thermal focal shift is less than the paraxial depth of focus. For example, an F/1.3 lens operating at a center wavelength of 10µm would have a theoretical depth of focus of ±34µm, so any Proc.